论文部分内容阅读
提出一种新的图像矢量量化编码算法．该算法结合相关矢量量化编码算法与地址矢量量化编码算法，首先根据相邻块状态进行编码地址的相关预测，对相关预测失败的图像块采用地址码书匹配编码，并采用自适应算术编码进行．测试结果表明，相对于无记忆矢量量化编码算法，比特率约下降３８％ —６８％ ，相对于安平提出的矢量量化地址编码算法以及王卫等提出的相关矢量量化编码算法，比特率约下降２５％ 以上
A new image vector quantization coding algorithm is proposed. The algorithm combines the vector quantization coding algorithm and the vector quantization algorithm of address vector. Firstly, the prediction of the coding address is carried out according to the state of the neighboring blocks. The code blocks with the corresponding prediction failure are coded using address code book matching, and adaptive arithmetic coding is used. The test results show that the bit rate is about 38% -68% lower than that of the memoryless vector quantization coding algorithm. Compared with the vector quantization address coding algorithm proposed by Anping and the related vector quantization coding algorithm proposed by Wang Wei, the bit rate is reduced by about 25% % the above